Despite impressive results with new therapies in laboratory settings, a major hurdle to their translation into the clinic is their suboptimal performances in vivo largely due to inefficient drug delivery in practical scenarios. A main challenge in optimizing drug delivery is to non-invasively quantify drug accumulation/internalization and/or monitor receptor dimerization within live subjects. Receptor dimerization can be systematically monitored in vivo with quantitative Frster Resonance Energy Transfer (FRET) imaging, enabled by highly-sensitive optical whole body imaging, compressive sensing and near-infrared (NIR) fluorescence lifetime FRET. Our proposed imaging strategy will combine these cutting-edge methodologies to tomographically depict protein-protein interactions in vivo. Under our current R21 support and based on strong academic partnership between the Rensselaer Polytechnic Institute (RPI) and the Albany Medical College (AMC), we have been developing state-of-the-art wide-field optical tomography instrumentation and algorithms for in vivo FRET imaging over the past 2 years. We have demonstrated the feasibility of quantitatively imaging FRET based on NIR fluorophore pair and lifetime sensing. To support the development of novel receptor-targeted therapy, the overall goal of this R01 is to develop and integrate key technologies for time-resolved wide-field molecular optical tomography and demonstrate its transformative ability to quantify receptor dimerization in live small animal models. With a hundred fold increase in sensitivity, improved resolution and quantification as well as accelerated in vivo imaging speed (<5minutes/frames), our proposed whole-body FRET imaging work promises to establish a new analytical tool with important and immediate drug development applications and beyond. This proposal synergistically integrates unique and powerful innovations for in vivo FRET imaging, specifically addressing the longstanding problems of practical implementation for time-resolved wide-field molecular optical tomography. High-spatial resolution and quantitative accuracy will be achieved with cutting-edge compressive sensing based reconstruction algorithms harnessing time-gate datasets. Fast acquisition will be attained in new protocols based on sparse temporal data patterns. Transferrin (Tfn) NIR FRET assays will be employed in a murine model bearing human breast cancer xenografts to establish lifetime-based FRET to quantify receptor dimerization and internalization. Moreover, lifetime-based FRET will be employed to identify non-invasively the Tfn-system with best cellular residency for improved drug delivery. Upon completion of the project, we will have demonstrated a breakthrough methodology for tomographic imaging of receptor dimerization, and hence direct imaging of receptor-mediated cellular internalization in tumors in small animals, offering unique powerful insight and guidance that would generate huge benefits for drug development in particular and healthcare at large.